Description of advanced parameters for job management - SchedulerX

This topic describes the advanced parameters for job management.

The following table describes the advanced parameters for job management.

Parameter	Applicable execution mode	Description	Default value
Task failure retry count	All modes	The number of automatic retries if a job fails. Note If a job is running on a worker and the worker is restarted, the job fails. If you want to immediately rerun the job, you can specify this parameter.	0
Task failure retry interval	All modes	The interval between two consecutive retries. Unit: second.	30
Task concurrency	All modes	The number of instances that run the same job at the same time. A value of 1 indicates that concurrent execution is not allowed.	1
Cleaning strategy	All modes	The cleanup policy for job execution history.	Keep last N entries
Retained Number	All modes	The number of retained job execution records.	300
Number of single-machine concurrent subtasks	Parallel computing Memory grid Grid computing	The number of tasks that concurrently run on a single worker in a distributed model.	5
Number of failed retries of subtasks	Parallel computing Memory grid Grid computing	The number of automatic retries if a task fails in a distributed model.	0
Sub-task failure retry interval	Parallel computing Memory grid Grid computing	The interval between two consecutive retries if a task fails in a distributed model. Unit: seconds.	0
Subtask distribution method	Parallel computing Memory grid Grid computing	Push model: Tasks are evenly distributed to workers. Pull model: Every worker pulls tasks. The Wooden Bucket Theory is not applicable to this model. During the pull process, all tasks are cached on the master node. This puts pressure on the memory. We recommend that you do not distribute more than 10,000 tasks at a time.	Push model
Number of subtasks pulled per time (applicable only if the Subtask distribution method parameter is set to Pull model)	Parallel computing Memory grid Grid computing	The number of tasks that a slave node pulls from the master node at a time.	5
Subtask queue capacity (applicable only if the Subtask distribution method parameter is set to Pull model)	Parallel computing Memory grid Grid computing	The size of the queue that caches tasks on a slave node.	10
Global concurrency of subtasks (applicable only if the Subtask distribution method parameter is set to Pull model)	Parallel computing Memory grid Grid computing	The total number of concurrent tasks on all workers in the pull model. This parameter helps you limit the number of concurrent tasks.	1,000